AI Song Recommender System

• Presented by: Tirth Jain (B021),

Mansi Kharke(B026),

Jash Damani(B013),

Dhruv Gupta(B019)

​

• Mentor: Prof. Mahesh Mourya

Computer Engineering Dept. MPSTME, Mumbai Campus

1

3/16/2022

Outline

Computer Engineering Dept. MPSTME, Mumbai Campus

2

3/16/2022

• Introduction
• Problem definition
• Literature review in brief (2 slides)
• Algorithms /Architecture
• UML Diagrams(Usecase, sequence,class diagram)
• Implementation
• Conclusion & Future wok
• References

​

​

​

Introduction

Song recommendations have existed for a long time, but in majority of the scenarios the recommendation is determined after learning the user preferences over a period of time, like looking at his past song preferences, time he listens to the music etc.

Computer Engineering Dept. MPSTME, Mumbai Campus

3

3/16/2022

Problem Statement

In this paper we propose a new approach to song recommendation, where in the mood of a person is determined from his picture and based on the mood predicted song recommendations are made that best suit the mood predicted.

Computer Engineering Dept. MPSTME, Mumbai Campus

4

3/16/2022

Literature Review

• Rapid development of mobile devices and internet has made possible for us to access different music resources freely. The number of songs available exceeds the listening capacity of single individual. People sometimes feel difficult to choose from millions of songs. Moreover, music service providers need an efficient way to manage songs and help their customers to discover music by giving quality recommendation.
• Thus, there is a strong need of a good recommendation system. Currently, there are many music streaming services, like Pandora, Spotify, etc. which are working on building high-precision commercial music recommendation systems. These companies generate revenue by helping their customers discover relevant music and charging them for the quality of their recommendation service.

Computer Engineering Dept. MPSTME, Mumbai Campus

5

3/16/2022

Literature Review

• Thus, there is a strong thriving market for good music recommendation systems. Music recommender system is a system which learns from the users past listening history and recommends them songs which they would probably like to hear in future. We have implemented various algorithms to try to build an effective recommender system.
• We firstly implemented popularity based model which was quite simple and intuitive. Collaborative filtering algorithms which predict (filtering) taste of a user by collecting preferences and tastes from many other users (collaborating) is also implemented. We have also done experiments on content based models, based on latent factors and metadata.

Computer Engineering Dept. MPSTME, Mumbai Campus

6

3/16/2022

Algorithm

We first process the image of the user taken as an input with the help of a python library for Computer Vision called 'OpenCV'. This captured image is then made available for the CNN in combination with DNN to make a prediction whether the current mood of the user is 'Happy' or 'Sad'.

​

The second part is the usage of Unsupervised Machine Learning techniques for clustering songs.The songs are clustered as either of the two classes-'VERY ENTERTAINING'(class 0) and 'RELAXED'(class 1) using the popular K-means algorithm. Then the recommendation is made in order of the current popularity of the respective songs.

​

Computer Engineering Dept. MPSTME, Mumbai Campus

7

3/16/2022

Algorithm

We have an unique story in the way we recommend the songs for each mood, for example when other sites recommend sad songs when a person is sad or feeling bad, we recommend users with songs which will cheer them up('VERY ENTERTAINING') and 'RELAXING' songs when they are 'HAPPY'.

​

The code to train the neural network can be found in the 'Emotion_detector_version2' iPython notebook. If anyone wants to modify the network to suit their particular needs or feel it is necessary to tweak the network they can do so by making changes to the code present there. The model created is saved as 'final_model.h5'

​

Computer Engineering Dept. MPSTME, Mumbai Campus

8

3/16/2022

UML Architecture

Computer Engineering Dept. MPSTME, Mumbai Campus

9

3/16/2022

Code Implementation

Computer Engineering Dept. MPSTME, Mumbai Campus

10

3/16/2022

Code Implementation

Computer Engineering Dept. MPSTME, Mumbai Campus

11

3/16/2022

Conclusion And Future Work

This is a project of our Artificial Intelligence course. We find it is very good as we got a chance to practice theories that we have learnt in the course, to do some implementation and to try to get a better understanding of a real artificial intelligence problem: Music Recommender System.

There are many different approaches to this problem and we get to know some algorithms in detail and especially the four models that we've explained in the paper. By manipulating the dataset, changing the learning set and testing set, changing some parameters of the problem and analyzing the result, we earn a lot practicing skills.

We've faced a lot of problem in dealing with this huge dataset, how to explore it in a better way and we also had difficulties in some programming details. However, with lot of efforts, we have overcame all of these.

​

​

Computer Engineering Dept. MPSTME, Mumbai Campus

12

3/16/2022

Conclusion And Future Work

• Run the algorithms on a distributed system, like Hadoop or Condor, to parallelize the computation, decrease the runtime and leverage distributed memory to run the complete MSD.
• Combine different methods and learn the weightage for each method according to the dataset
• Automatically generate relevant features
• Develop more recommendation algorithms based on different data (e.g. the how the user is feeling, social recommendation, etc)

​

​

Computer Engineering Dept. MPSTME, Mumbai Campus

13

3/16/2022

References

https://cyanite.ai/2021/09/02/how-do-ai-music-recommendation-systems-work/

​

https://cse.iitk.ac.in/users/cs365/2014/_submissions/shefalig/project/report.pdf

​

Computer Engineering Dept. MPSTME, Mumbai Campus

14

3/16/2022